Kontogianni, Afroditi; Galanaki, Evangelia (2014). 'Anxiety symptoms in school-age children in relation to bullying and victimization' Paper presented at the annual conference of the HDCA, 2-5 September 2014, Athens, Greece.

The purpose of this study was to examine the relation between school bullying/victimization and symptoms of anxiety disorders in school-age children. Social anxiety (Slee, 1994; Storch, Brassard, & Masia-Warner, 2003), post-traumatic stress disorder (Idsoe, Dyregrov, & Cosmovici Idsoe, 2012; Mynard, Joseph, & Alexander, 2000), as well as psychosomatic symptoms (Baldry, 2004; Gini, 2008) have all been found to be associated with bullying and victimization in schools. The psychological well-being of students involved in bullying incidents is a rather neglected research issue in Greece (Andreou, 2000), whereas prevalence or the various forms of bullying have been adequately discussed. Therefore, the main goal of this study was to identify the possible connections between school bullying and victimization and the occurrence of anxiety disorders in preadolescent children. Moreover, the role of the child in a bullying incident, that is, bully, victim, bully-victim or non bully-non victim, was examined in relation to anxiety disorders.

Participants were 255 fifth- and sixth-grade students (mean age = 12 years) from primary schools of Attica region, Greece. The majority of the students were born in Greece (94.9%), thus they could comprehend and write the Greek language. They completed the Greek versions of the following measures: the Revised Olweus Bully/Victim Questionnaire (Galanaki & Amanaki, 2009; Olweus, 2006), the Social Anxiety Scale for Children-Revised (SASC-R; LaGreca & Stone, 1993; Vassilopoulos, 2009), the Children's Revised Impact of Event Scale (CRIES-13; Giannopoulou, Smith, Ecker, Strouthos, Dikaiakou, & Yule, 2006; Smith, Perrin, Dyregrov, & Yule, 2003) and the Psychosomatic Symptom Score (PSS; Aro, Paronen, & Aro, 1987; the scale was adapted in Greek by the researcher).

The study showed that the percentage of victims (8.6%) was in accordance with that found in other studies, followed by lower percentages for bullies (2.4%) and bully-victims (2.0%). The most common forms of victimization were verbal victimization (12.6%), social exclusion (7.4%) and sexual victimization (7.1%). Accordingly, the most frequent forms of bullying were verbal bullying (5.9%), social exclusion (3.2%) and physical bullying (2.4%). The mean score of the tendency to be bullied was 1.28, while the mean score of the tendency to bully others was 1.12 (at a 5-point Likert scale).

Results showed that both the tendency to be bullied and the tendency to bully others correlated significantly with all measures of anxiety, although the correlations were higher for the tendency to be bullied. As the tendency to be bullied and the tendency to bully others increase, so do general social anxiety, fear of negative evaluation (FNE), social avoidance and distress-specific to new peers or situations (SAD-New), and the social avoidance and distress-general (SAD-General). A positive association of the tendency to be bullied and the tendency to bully others with psychosomatic symptoms was also found. Post-traumatic stress disorder was also positively correlated to the tendency to be bullied and the tendency to bully others, and this was found both for the total scale score and its subscales (Avoidance, Intrusion, Arousal).

Multiple linear regression analysis was performed using the total scores of the scales. Students at higher risk for post-traumatic stress disorder showed significantly higher tendency to be bullied in comparison to other students. The more the tendency to be bullied increased, the more students complained of psychosomatic symptoms and manifested clinically significant social anxiety. The subscales of each anxiety scale were then used in multiple linear regression analysis. Psychosomatic problems again predicted the tendency to be bullied. Moreover, the FNE and the SAD-New subscales were independently linked to the tendency to be bullied: as FNE and SAD-S increased, so did the tendency to be bullied. Only the subscale Intrusion of the PTSD scale was independently connected to the tendency to be bullied.

Multiple linear regression analysis was conducted for the tendency to bully others as well. Using the total scores of the scales, only psychosomatic symptoms were found to be related to the students' tendency to bully others. Moreover, students with mothers born in Greece had significantly lower tendency to bully others, when compared to students with mothers born in other countries. Subsequently, the subscales of each anxiety scale were used in multiple linear regression analysis. Psychosomatic problems and mothers who were not born in Greece again predicted a higher tendency to bully others. Additionally, as FNE and SAD-General of the social anxiety and Arousal of the PTSD scale increased, so did the tendency to bully others.

Compared to non bullies-non victims, victims were found to score higher on all the anxiety measures (i.e., all the subscales of social anxiety, PTSD and the psychosomatic problems). Multivariate logistic regression (using odds ratio) showed that students with higher social anxiety and higher PTSD were more likely to be victimized. Using the subscales, FNE of the social anxiety scale and Avoidance of the PTSD scale were more likely to be reported by students who admitted they were victims of school bullying.

The study reveals the need for further research on students' well-being with regard to bullying problems in their schools. In order for children to obtain a higher level of well-being, educational policies must provide all schools with the same educational primary goods, offering the same opportunities to all students and implementing appropriate programs. Bullying is mainly a social phenomenon; consequently, when we are able to adequately address it, we facilitate children realize their full potential and thrive.